In this edition of Web Enlightenment we'll be talking about how data is simply the first step in analysis.

This newsletter focuses on giving you practical, interesting insights into how to successfully use technology as a tool to improve the way you do business. Bridging the gap between knowledge and understanding that all entrepreneurs have will help you make money online.

Please hit reply and tell me about a piece of web data that confuses you. I answer every email sent to me.

All the best,

Data is not Analysis

I love talking with people about geeky things - it is just the most fun.

Recently I've been speaking to a wide variety of groups and I keep running across what I call the "oh" moment, like when I gave a presentation called "Web Analytics Demystified" to the folks at the Mailorder Gardening Association.

One of the things I discussed was the technical methods used to calculate web statistics and I explained "The Hotel Problem". This is a great way of showing why the number of unique visitors per day added together for 7 days do not equal the number of unique visitors per week.

I love the expression on people's faces when you point out that 2+2 does not equal four and that 2+2+2 does not equal six - the answer to both questions is actually three. I may well be confusing you senseless right here and if I am feel free to email me and I will send you a handout on this issue - but today we're talking about something a bit different.

Most Internet Entrepreneurs receive information about their website as raw data. That's a big problem, because very few of them have the training required to read the data and turn it into useful actions, which is the reason why we collect the data in the first place.

I think this is because for many years there has been so much focus on tracking. The majority of articles on this subject ask questions like:

• Are you tracking your site ?
• What system/software do you use to track your site ?
• Are you aware of the latest innovations in tracking systems and do you have them ?

And all of this can easily lead you to a spot where you congratulate yourself on the fact you have systems, that reports are being generated and leave you with the feeling you've got it all covered.

That's dangerous.

A very small number of web metrics are readable by anyone and they make sense easily, I like to call these level one metrics.

• Visits - more people coming to your site is good, it is like the circulation of your ad.
• Referrers - where people came from (especially if you match it up with a list of where you spent ad dollars) is good and it can help you decide which advertising to do more of and which to kill.
• Popular Pages - This is a great indicator of what is important to your customers and what you should work on next (does our "about us" page even matter?)

But then - very quickly - you get into the next level of metrics, level two. At that level we run into things where there is more than one way to look at the data, like bounce rate. A "bounce" is when someone comes to your site and immediately leaves again. It is often used as an indicator metric for design quality in the same way we'd think about a display window in a brick and mortar store and whether or not that causes people to come inside. Most Internet Entrepreneurs consider bounces "bad" and they want to reduce the bounce rate.

But you know what? Adding your Toll Free number big and bold above the fold will reliably increase your bounce rate. And that's wonderful - because it means customers are coming to your site looking for your phone number and they are finding it right away, a wonderful experience for them.

So how do you tell the difference between a "bounce" where a person found what they wanted right away and one where they decided your site sucks and they were going to go back to Google? You can't. It is impossible to discern.

And then we have the third level of web metrics - where things work in tandem and have relationships of their own. A huge boost in traffic always reduces your conversion rate, adding more pages to your site always increases the pages/visits number - it is a big problem when you try to look at one of these level three metrics in isolation. That almost always leads the Internet Entrepreneur to make "conclusions" that are just plain wrong.

So bottom line what do you need to know about all of this?

The best analysis reports for site owners actually contain very little raw data, sometimes they contain none at all. Knowing that site visitation is up 18 % or down 5 % is a heck of a lot more important than knowing the raw number.

When you deal with the geeky person that provides this information to you you need to make sure they understand that you have three questions.

• What happened?
• What did we learn?
• What should we do now? (far and away the most important question)

This will give you a wonderful method by which you can know nothing about web metrics or how they work and yet still make sure that they are working for your organization. You take the first "what we learned" and "what we should do now" on faith, but in the next reporting period it is pretty simple for you to find out whether or not what we did worked.

As an example, suppose we think people consider "about us" information to be really important so we're going to expand that section. Ok, so with that done did it increase people converting from that page? Did more people visit the expanded page(s)? Did they spend more time there?

As an Internet Entrepreneur you want to make sure you are "closing the loop" by holding folks accountable over time to constantly improve the site. This is best done with reports that contain almost no raw data and that allow you to judge the effectiveness of activity from reporting period to reporting period.

MANP Freebies

Congratulations to Allison Cole of Hardy Girls Healthy Women and Bob Holtzman of Your Name Here Communications.

They were the first two people to reply to my last newsletter so they received the free passes to an upcoming MANP event.